U.S. patent application number 10/973029 was filed with the patent office on 2006-05-11 for network modeling systems and methods.
Invention is credited to Jiang Li.
Application Number | 20060101135 10/973029 |
Document ID | / |
Family ID | 36317641 |
Filed Date | 2006-05-11 |
United States Patent
Application |
20060101135 |
Kind Code |
A1 |
Li; Jiang |
May 11, 2006 |
Network modeling systems and methods
Abstract
Embodiments of network modeling systems and methods are
disclosed. In one method embodiment, the network modeling method
includes receiving multiple 4-port s-parameter measurements
corresponding to an 8-port device and generating an 8-port model
from the multiple 4-port s-parameter measurements.
Inventors: |
Li; Jiang; (Plano,
TX) |
Correspondence
Address: |
HEWLETT PACKARD COMPANY
P O BOX 272400, 3404 E. HARMONY ROAD
INTELLECTUAL PROPERTY ADMINISTRATION
FORT COLLINS
CO
80527-2400
US
|
Family ID: |
36317641 |
Appl. No.: |
10/973029 |
Filed: |
October 25, 2004 |
Current U.S.
Class: |
709/223 |
Current CPC
Class: |
G06F 30/20 20200101 |
Class at
Publication: |
709/223 |
International
Class: |
G06F 15/173 20060101
G06F015/173 |
Claims
1. A network modeling method, comprising: receiving six 4-port
s-parameter measurements corresponding to an 8-port device; saving
the six 4-port s-parameter measurements in a plurality of data
files; and combining the plurality of data files into a model data
file, the model data file representing the 8-port device.
2. The method of claim 1, further including acquiring the six
4-port s-parameter measurements from a measurement device having 4
ports from which the s-parameter measurements are taken.
3. The method of claim 2, wherein the measurement device includes a
vector network analyzer.
4. The method of claim 1, wherein saving includes saving in a text
file.
5. The method of claim 1, wherein combining includes executing a
postscript operation on the plurality of data files.
6. The method of claim 1, wherein the model data file includes
information corresponding to at least one of far end cross talk,
near end cross talk, and pass through.
7. The method of claim 1, wherein the 8-port device includes a
device under test.
8. The method of claim 7, wherein the device under test includes at
least one of a network, a component, and a signal path.
9. The method of claim 1, further including providing the model
data file to simulation software to be executed to characterize the
performance of the 8-port device.
10. A network modeling system, comprising: memory with modeling
software; and a processor configured with the modeling software to
receive multiple 4-port s-parameter measurements corresponding to
an 8-port device, save the multiple 4-port s-parameter measurements
in a plurality of data files, and combine the plurality of data
files into a model data file, the model data file representing the
8-port device.
11. The system of claim 10, wherein the processor is configured
with the modeling software to receive six 4-port s-parameter
measurements corresponding to an 8-port device and save the six
4-port s-parameter measurements in a plurality of data files.
12. The system of claim 10, further including a measurement device,
wherein the measurement device includes 4 ports from which the
s-parameter measurements are taken.
13. The system of claim 12, wherein the measurement device includes
a vector network analyzer.
14. The system of claim 10, wherein the 8-port device includes a
device under test, the device under test configured with at least
one of a signal path to be measured and a component having
predetermined performance features.
15. The system of claim 10, wherein the processor is configured
with the modeling software to execute a postscript operation on the
plurality of data files.
16. The system of claim 10, wherein the processor is configured
with the modeling software to save the plurality of data files in
respective text files.
17. The system of claim 10, wherein the model data file includes
information corresponding to at least one of far end cross talk,
near end cross talk, and pass through for the 8-port device.
18. The system of claim 10, wherein the modeling software is
included in at least one of a computer and a vector network
analyzer.
19. A network modeling system, comprising: means for receiving six
4-port s-parameter measurements corresponding to an 8-port device;
means for saving the six 4-port s-parameter measurements in a
plurality of data files; and means for combining the plurality of
data files into a model data file, the model data file representing
the 8-port device.
20. The system of claim 19, wherein the means for receiving,
saving, and combining includes software in memory, the software
executed by a processor.
21. A computer program for modeling a network, the program being
stored on a computer-readable medium, the computer-readable medium
comprising: logic configured to receive multiple 4-port s-parameter
measurements corresponding to an 8-port device; logic configured to
save the multiple 4-port s-parameter measurements in a plurality of
data files; and logic configured to combine the plurality of data
files into a model data file, the model data file representing the
8-port device.
22. A network modeling method, comprising: generating a plurality
of 8-port models, each of the plurality of 8-port models
corresponding to a victim pair and a culprit pair for a multi-port
device, the multi-port device having N ports; and combining the
plurality of 8-port models to generate an N-port model.
23. The method of claim 22, wherein generating a plurality of
8-port models includes generating a plurality of data files and
combining the plurality of data files into a model data file, the
model data file representing an 8-port device.
24. The method of claim 22, wherein combining includes combining a
plurality of 8-port model data files.
25. The method of claim 22, further including determining a
quantity of culprit pairs in the N-port model.
26. The method of claim 22, further including determining whether
an 8-port model has been generated that includes the victim pair
and every culprit pair.
27. A network modeling system, comprising: a memory with modeling
software; and a processor configured with the modeling software to
generate a plurality of 8-port models, each of the plurality of
8-port models corresponding to a victim pair and a culprit pair for
an N-port device, wherein the processor is configured with the
modeling software to combine the plurality of 8-port models to
generate an N-port model.
28. The system of claim 27, wherein the processor is configured
with the modeling software to generate a plurality of data files
corresponding to s-parameter measurements of the N-port device and
combine the plurality of data files into a model data file, the
model data file representing an 8-port device.
29. The system of claim 27, wherein the processor is configured
with the modeling software to combine a plurality of 8-port model
data files.
30. The system of claim 27, wherein the processor is configured
with the modeling software to determine a quantity of culprit pairs
in the N-port model.
31. The system of claim 27, wherein the processor is configured
with the modeling software to determine whether an 8-port model has
been generated that includes the victim pair and every culprit
pair.
32. The system of claim 27, wherein the modeling software is
included in at least one of a computer and a vector network
analyzer.
33. A network modeling system, comprising: means for generating a
plurality of 8-port models, each of the plurality of 8-port models
corresponding to a victim pair and a culprit pair for a multi-port
device, the multi-port device having N ports; and means for
combining the plurality of 8-port models to generate an N-port
model.
34. The system of claim 33, wherein the means for generating and
combining includes software in memory, the software executed by a
processor.
35. A computer program for modeling a network, the program being
stored on a computer-readable medium, the computer-readable medium
comprising: logic configured to generate a plurality of 8-port
models, each of the plurality of 8-port models corresponding to a
victim pair and a culprit pair for a multi-port device, the
multi-port device having N ports; and logic configured to combine
the plurality of 8-port models to generate an N-port model.
36. A network modeling method, comprising: receiving multiple
4-port s-parameter measurements corresponding to an 8-port device;
and generating an 8-port model from the multiple 4-port s-parameter
measurements.
Description
BACKGROUND
[0001] Network modeling is a technique often used to represent
physical components, signal paths, and/or systems in general. For
instance, designers of proposed network topologies, such as for a
semiconductor circuit design, often use one or more models to
characterize signal paths. The model can then be used in
simulations, using various design software such as SPICE, which
provides for observation of performance and enables designers and
other persons to make decisions on component and/or system design
choice. One approach that may be used to model a signal path
includes building the actual hardware and testing it. However, an
often less expensive approach is to build a model out of various
components of the proposed network topology, and simulate outputs
under various input scenarios. This may also be the only feasible
approach when system hardware is not available for testing.
[0002] For networks such as high-speed digital links, current
models may pose limitations. For example, RLC
(resistor-inductor-capacitor) models are typically implemented by a
user inputting a signal path structure using a limited data format.
Further, the assumptions and/or simplifications of RLC models as
well as the analysis engine/methodology often limit accuracy. The
fact that RLC models are static tools also limits their
effectiveness at high data rates.
[0003] Measurement-based models may provide an improvement over RLC
models. For example, a device under test (DUT) may be configured
with various components that provide a variety of signal paths
(thus providing a multitude of measurable signal performance
characteristics). High data rates can typically be accommodated in
measurement-based models. However, measurement-based models may be
limited by the equipment available, among other limitations. For
instance, measurement equipment currently available generally
includes one single-ended signal path (e.g., 2-port) or one
differential signal path (e.g., 4-port). With limited port
availability, measurement-based models may fail to include some
information that is important to network design, such as cross-talk
information, or may be hindered for networks that are represented
using more than the amount of ports available on the measurement
equipment.
SUMMARY
[0004] An embodiment of a network modeling method comprises
receiving six 4-port s-parameter measurements corresponding to an
8-port device; saving the six 4-port s-parameter measurements in a
plurality of data files; and combining the plurality of data files
into a model data file, the model data file representing the 8-port
device.
[0005] An embodiment of a network modeling system comprises a
memory with modeling software; and a processor configured with the
modeling software to receive multiple 4-port s-parameter
measurements corresponding to an 8-port device, save the multiple
4-port s-parameter measurements in a plurality of data files, and
combine the plurality of data files into a model data file, the
model data file representing the 8-port device.
[0006] An embodiment of a network modeling system comprises means
for receiving six 4-port s-parameter measurements corresponding to
an 8-port device; means for saving the six 4-port s-parameter
measurements in a plurality of data files; and means for combining
the plurality of data files into a model data file, the model data
file representing the 8-port device.
[0007] An embodiment of a computer program for modeling a network,
the program being stored on a computer-readable medium, comprises
logic configured to receive multiple 4-port s-parameter
measurements corresponding to an 8-port device; logic configured to
save the multiple 4-port s-parameter measurements in a plurality of
data files; and logic configured to combine the plurality of data
files into a model data file, the model data file representing the
8-port device.
[0008] An embodiment of a network modeling method comprises
generating a plurality of 8-port models, each of the plurality of
8-port models corresponding to a victim pair and a culprit pair for
a multi-port device, the multi-port device having N ports; and
combining the plurality of 8-port models to generate an N-port
model.
[0009] An embodiment of a network modeling system comprises a
memory with modeling software; and a processor configured with the
modeling software to generate a plurality of 8-port models, each of
the plurality of 8-port models corresponding to a victim pair and a
culprit pair for an N-port device, wherein the processor is
configured with the modeling software to combine the plurality of
8-port models to generate an N-port model.
[0010] An embodiment of a network modeling system comprises means
for generating a plurality of 8-port models, each of the plurality
of 8-port models corresponding to a victim pair and a culprit pair
for a multi-port device, the multi-port device having N ports; and
means for combining the plurality of 8-port models to generate an
N-port model.
[0011] An embodiment of a computer program for modeling a network,
the program being stored on a computer-readable medium, comprises
logic configured to generate a plurality of 8-port models, each of
the plurality of 8-port models corresponding to a victim pair and a
culprit pair for a multi-port device, the multi-port device having
N ports; and logic configured to combine the plurality of 8-port
models to generate an N-port model.
[0012] An embodiment of a network modeling method comprises
receiving multiple 4-port s-parameter measurements corresponding to
an 8-port device; and generating an 8-port model from the multiple
4-port s-parameter measurements.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The components in the drawings are not necessarily to scale,
emphasis instead being placed upon clearly illustrating the
principles of the disclosed systems and methods. Moreover, in the
drawings, like reference numerals designate corresponding parts
throughout the several views.
[0014] FIG. 1A is a block diagram that illustrates an embodiment of
a network modeling system.
[0015] FIG. 1B is a block diagram that illustrates an embodiment of
a computer configured with modeling software as shown in FIG.
1A.
[0016] FIG. 2 is a flow diagram of a method embodiment of the
modeling software shown in FIG. 1B, the method providing for
modeling or representing an 8-port network using six 4-port
s-parameter measurements.
[0017] FIGS. 3A-3F are schematic diagrams of exemplary port
configurations used to obtain s-parameter measurements that fully
characterize an 8-port network using the modeling method shown in
FIG. 2.
[0018] FIGS. 4A-4F are schematic diagrams that illustrate matrix
processing as implemented by the method shown in FIG. 2, the
matrices generated based on the port configurations shown in FIGS.
3A-3F.
[0019] FIG. 5 is a flow diagram of a method embodiment of the
modeling software shown in FIG. 1B, the method providing for
modeling cross-talk for multi-port networks.
[0020] FIG. 6 is a schematic diagram that illustrates a 12-port
device under test (DUT) with a victim pair and two culprit
pairs.
[0021] FIGS. 7A-C are schematic diagrams that illustrate matrix
processing as implemented by the method shown in FIG. 5, the
matrices generated based on 8-port network modeling shown in FIGS.
1A-4F.
[0022] FIG. 8 is a schematic diagram that illustrates a 16-port DUT
with a victim pair and three culprit pairs.
[0023] FIGS. 9A-9D are schematic diagrams that illustrate matrix
processing as implemented by the method shown in FIG. 5, the
matrices generated based on 8-port network modeling shown in FIGS.
1A-4F.
DETAILED DESCRIPTION
[0024] Disclosed are various embodiments of network modeling
systems and methods (herein referred to as a network modeling
system for brevity). In one embodiment, a network modeling system
includes functionality to characterize the behavior of (i.e., to
model or represent) an 8-port network using six, 4-port s-parameter
analyzer measurements. The resulting network model can be used in
simulations to characterize the electrical performance of
high-speed links, with bandwidths generally ranging from DC to 20
giga-Hertz (GHz). A network modeling system also includes
functionality to characterize multi-port networks beyond an 8-port
network (e.g., 12-ports, 16-ports, etc.), providing a frequency
domain differential cross-talk model for high-speed links.
[0025] S-parameters (or scattering parameters) generally refer to
reflection and transmission coefficients between incident and
reflection signals, and can be used to describe the behavior of a
device. Also, a link generally refers to a communication medium
between components, such as a signal path between two ASICs
(application specific integrated circuits).
[0026] An embodiment of a network modeling system is illustrated in
FIG. 1A, which includes a computer configured with modeling
software in communication with a network analyzer that acquires
s-parameter measurements from a device under test (DUT). FIG. 1B
illustrates a computer architecture embodiment, and FIG. 2 shows a
method embodiment of the modeling software. FIGS. 3A-3F illustrate
various port configurations used to take s-parameter measurements
from a DUT configured with 8-ports. FIGS. 4A-4F illustrate matrix
processing implemented by the modeling software to fully
characterize the 8-port DUT. FIG. 5 illustrates a modeling method
embodiment that characterizes coupling (e.g., cross-talk) in
multi-port networks, and FIGS. 6-9D provide 12-port and 16-port
coupling illustrations and matrix processing for the same. It will
be understood that the principles disclosed herein can be applied
to multi-port devices and networks in addition to the disclosed
examples.
[0027] FIG. 1A is a block diagram that illustrates an embodiment of
a network modeling system 100. The network modeling system 100
includes an exemplary vector network analyzer (VNA) 102, a device
under test (DUT) 106, and a computer 120. The VNA 102 includes four
front panel ports 104 (labeled 1-4). The VNA 102 takes s-parameter
measurements of the DUT 106 using a plurality of connection
configurations 105, as described below. The VNA 102 may display the
s-parameter measurements in a curve or format the same in one or
more data files. The DUT 106 may represent one or more devices, the
signal paths between and/or including the devices, or a network.
Although shown with four connections, the DUT 106 can have a
different quantity of connections. The computer 120 includes
modeling software 110. The modeling software 110 receives the
s-parameter measurements from the VNA 102 and generates a
multi-port network model. Although shown using a vector network
analyzer 102, other measurement/diagnostic equipment may be
used.
[0028] FIG. 1B is a block diagram that illustrates an embodiment of
the computer 120. The computer 120 includes the modeling software
110 that receives s-parameterization measurements and configures
measurements into an 8-port network model. The modeling software
110, or like-functionality, can be implemented in whole or in part
in the computer 120, or in some embodiments, in other devices such
as the VNA 102. The computer 120 may include fewer or additional
components. Generally, in terms of hardware architecture, the
computer 120 includes a processor 160, memory 158, and one or more
input and/or output (I/O) devices 170 that are communicatively
coupled via a local interface 180. The local interface 180 can be,
for example but not limited to, one or more buses or other wired or
wireless connections. The local interface 180 may have additional
elements, which are omitted for simplicity, such as controllers,
buffers (caches), drivers, repeaters, and receivers, to enable
communications. Further, the local interface 180 may include
address, control, and/or data connections to enable appropriate
communications among the aforementioned components.
[0029] The processor 160 is a hardware device for executing
software, particularly that which is stored in memory 158. The
processor 160 can be any custom made or commercially available
processor, a central processing unit (CPU), an auxiliary processor
among several processors associated with the computer 120, a
semiconductor-based microprocessor (in the form of a microchip or
chip set), a macroprocessor, or generally any device for executing
software instructions.
[0030] Memory 158 can include any one or combination of volatile
memory elements (e.g., random access memory (RAM, such as DRAM,
SRAM, SDRAM, etc.)) and nonvolatile memory elements (e.g.,
read-only memory (ROM)). Memory 158 cooperates through the local
interface 180. In some embodiments, memory 158 may incorporate
electronic, magnetic, optical, and/or other types of storage media.
Note that memory 158 can have a distributed architecture, where
various components are situated remote from one another, but can be
accessed by the processor 160.
[0031] The software in memory 158 may include one or more separate
programs, each of which comprises an ordered listing of executable
instructions for implementing logical functions. In the example of
FIG. 1B, the software in memory 158 includes a suitable operating
system (O/S) 156, the modeling software 110, and simulation
software 114 (e.g., SPICE). In general, the operating system 156
essentially controls the execution of other computer programs, and
provides scheduling, input-output control, file and data
management, memory management, and communication control and
related services.
[0032] The modeling software 110 is a source program, executable
program (object code), script, or any other entity comprising a set
of instructions to be performed. The modeling software 110 can be
implemented as a single module or as a distributed network of
modules of like-functionality. When the modeling software 110 is a
source program, then the program is translated via a compiler,
assembler, interpreter, or the like, which may or may not be
included within the memory 158, so as to operate properly in
connection with the O/S 156.
[0033] The I/O devices 170 may include input devices, for example
but not limited to, a keyboard, mouse, scanner, microphone, etc.
Furthermore, the I/O devices 170 may also include output devices,
for example but not limited to, a printer, display, etc. Finally,
the I/O devices 170 may further include devices that communicate
both inputs and outputs, for instance but not limited to, a
modulator/demodulator (modem; for accessing another device, system,
or network), a radio frequency (RF) or other transceiver, a
telephonic interface, a bridge, a router, etc.
[0034] When the computer 120 is in operation, the processor 160 is
configured to execute software stored within the memory 158, to
communicate data to and from the memory 158, and to generally
control operations of the computer 120 pursuant to the software.
For example, the modeling software 110, in whole or in part, is
read by the processor 160, perhaps buffered within the processor
160, and then executed.
[0035] When the modeling software 110 is implemented in software,
as is shown in FIG. 1B, it should be noted that the modeling
software 110 can be stored on any computer-readable medium for use
by or in connection with any computer related system or method. In
the context of this document, a computer-readable medium is an
electronic, magnetic, optical, or other physical device or means
that can contain or store a computer program for use by or in
connection with a computer related system or method. The modeling
software 110 can be embodied in any computer-readable medium for
use by or in connection with an instruction execution system,
apparatus, or device, such as a computer-based system,
processor-containing system, or other system that can fetch the
instructions from the instruction execution system, apparatus, or
device and execute the instructions. The computer-readable medium
may be portable.
[0036] Any process descriptions or blocks in flow diagrams used
herein should be understood as representing modules, segments, or
portions of code which include one or more executable instructions
for implementing specific logical functions or steps in the
process, and alternate implementations are included within the
scope of the disclosure in which functions may be executed out of
order from that shown or discussed, including substantially
concurrently or in reverse order, depending on the functionality
involved.
[0037] With continued reference to FIGS. 1A-1B, FIG. 2 is a flow
diagram of a method embodiment 110a for the modeling software 110
shown in FIG. 1B. The modeling method 110a provides an 8-port
network model from six 4-port device s-parameter measurements. The
VNA 102 takes six 4-port s-parameter measurements from the DUT 106
and forwards the same to the modeling software 110 (202). The
modeling software 110 saves each of the six s-parameter
measurements in six individual s-parameter data files (204). The
modeling software 110 combines the six individual files into one
data file (206). The data file may be formatted for use by the
simulation software 114. In one embodiment, the individual data
files include a 4.times.4 matrix of s-parameter data that are
executed in a postscript process to generate the single data file.
The single data file may include an 8.times.8 matrix of the
measured s-parameters. Other mechanisms to combine the data files
may be implemented. The single data file represents or
characterizes the 8-port DUT 106. The 8-port DUT model (e.g.,
network model) is then used by the simulation software 114 to
provide performance characteristics for a particular network based
on a plurality of different inputs, signal paths, and/or
components.
[0038] FIGS. 3A-3F are schematic diagrams of exemplary port
connection configurations (105a-105f). A 4-port network is fully
characterized when information corresponding to 16 s-parameters
(e.g., s.sub.11, s.sub.12, etc.) are obtained. An 8-port network is
fully characterized when 64 s-parameters are obtained. For example,
to fully characterize near end cross-talk, far-end cross-talk, and
pass-through (i.e., unaffected by coupling), an 8.times.8 matrix of
s-parameter measurements are obtained by combining 6-4.times.4
matrices of s-parameter measurements. The exemplary port connection
configurations (105a-105f) enable the acquisition of all 64
s-parameters from six 4-port measurements to fully characterize the
8-port DUT 106. Sometimes, fewer measurements may be implemented in
some embodiments to achieve acceptable accuracy in the model.
Referring to FIG. 3A, a port connection configuration 105a is shown
that includes the four ports 104 of the VNA 102 (FIG. 1A) and the
DUT 106 shown with 8-ports (labeled port1-port8). Unused ports of
the DUT 106 (e.g., port3, port4, port7, and port8) are terminated
using, for example, a 50 .OMEGA. resistor (R) 112. FIGS. 3B-3F are
not shown with the resistors 112 for clarity, although it will be
understood that those DUT ports in configurations 105b-105f that
are shown without a connection to the VNA ports 104 are terminated
by the resistors 112.
[0039] In the set-up 105a shown in FIG. 3A, VNA port 1 and port 3
are connected to the DUT 106 at port2 and port1, respectively. VNA
port 2 and 4 are connected at the DUT at port5 and port6,
respectively.
[0040] Referring to the set-up 105b in FIG. 3B, VNA port 1 and port
3 are connected to the DUT 106 at port2 and port1, respectively.
VNA port 2 and 4 are connected at the DUT at port3 and port4,
respectively.
[0041] Referring to the set-up 105c in FIG. 3C, VNA port 1 and port
3 are connected to the DUT 106 at port2 and port1, respectively.
VNA port 2 and 4 are connected at the DUT at port7 and port8,
respectively.
[0042] Referring to the set-up 105d in FIG. 3D, VNA port 1 and port
3 are connected to the DUT 106 at port6 and port5, respectively.
VNA port 2 and 4 are connected at the DUT at port7 and port8,
respectively.
[0043] Referring to the set-up 105e in FIG. 3E, VNA port 1 and port
3 are connected to the DUT 106 at port4 and port3, respectively.
VNA port 2 and 4 are connected at the DUT at port7 and port8,
respectively.
[0044] Referring to the set-up 105f in FIG. 3F, VNA port 1 and port
3 are connected to the DUT 106 at port4 and port3, respectively.
VNA port 2 and 4 are connected at the DUT at port5 and port6,
respectively.
[0045] FIGS. 4A-4F are schematic diagrams that illustrate matrix
processing as implemented by the method 110a shown in FIG. 2, the
matrices generated based on the port configurations (105a-105f)
shown in FIGS. 3A-3F. Referring to FIG. 4A, an 8.times.8 matrix
405a is shown corresponding to the s-parameter measurements taken
with the configuration 105a (FIG. 3A). Each entry 401 in the matrix
includes an s-parameter element. Although shown using numerals only
(e.g., "13" in entry 401), it will be understood that each entry
corresponds to an s-parameter entry, such as "s.sub.13" for entry
401. In other words, "13" represents the s-parameter (s.sub.13)
measured when an input is provided at port3 of the DUT 106 and an
output is measured at port1 of the DUT 106. Shaded areas 403
represent which s-parameters are covered or measured for the
corresponding port connection configuration.
[0046] FIGS. 4B through 4F include matrices 405b-405f, which in
turn correspond to s-parameter measurements taken using port
connection configurations 105b-105f, respectively. For instance,
matrix 405b corresponds to the s-parameter measurements taken using
the port connection configuration 105b (FIG. 3B), and matrix 405c
corresponds to the s-parameter measurements taken using the port
connection configuration 105c (FIG. 3C), and so on.
[0047] A 4-port network is fully characterized when all 16
s-parameters (11-44) are measured, and an 8-port network is fully
characterized when all 64 s-parameters (11-88) are measured. Thus,
one goal is to cover (e.g., through measurement) all 64
s-parameters in an 8.times.8 matrix, such as shown in the matrix
405f in FIG. 4F, in which all s-parameters are covered (as
represented by the shading). In one embodiment, this is achieved by
taking six 4-port s-parameter measurements as explained above.
[0048] The above methodology to generate 8-port network models can
be applied to generate network models that include information
about cross-talk from different signal paths. Such models are
generally referred hereinafter as victim/culprit coupling models. A
victim generally refers to an intended signal path of a network or
device. A culprit generally refers to a signal path that corrupts
the victim, such as when high speed data wiring is bundled closely
together. In one embodiment, a victim/culprit coupling model may be
based on two or more frequency domain, 8-port differential
cross-talk models to evaluate the cross-talk from different culprit
pairs. Each of the 8-port models can be generated from the modeling
method 110a using the same victim signal pairs but different
culprits pairs (FIG. 2). Like the 8-port models described above, a
victim/culprit coupling model can be used (by the simulation
software 114) to characterize the electrical performance of
high-speed links.
[0049] FIG. 5 is a flow diagram of a method embodiment 110b of the
modeling software 110 (FIG. 1B), which provides for modeling
cross-talk for multi-port networks. In one embodiment, a
determination is made as to the number of culprit pairs, N (502).
Depending on the topology of the network, data rates, and packaging
(e.g., wiring proximity), there may be one or more culprit pairs.
The modeling software 110 generates an 8-port model with one victim
pair and one culprit pair (504). The 8-port generation occurs in a
manner as described in the method 110a illustrated in FIG. 2. If
there is more than one culprit pair (506), then an 8-port network
model is generated with the victim pair as determined above and a
second culprit pair (504). This process (504, 506, 508, 504, etc.)
repeats itself for each culprit pair up to N. When 8-port models
have been generated for N culprit pairs (including the victim pair
in each model), the 8-port models (e.g., the data files
corresponding to the s-parameter measurements) are combined to
create a multi-port model (510).
[0050] FIG. 6 is a schematic diagram that illustrates a 12-port DUT
606 with a victim pair 602 and two culprit pairs 603 and 604. Ports
are designated port1 through port12. With continued reference to
FIG. 6, FIGS. 7A-C are schematic diagrams of matrices 700a-700c,
respectively, that illustrate matrix processing as implemented by
the method 110b shown in FIG. 5. Regarding the matrix 700a of FIG.
7A, shaded portions, such as shaded portion 702a, represent
s-parameter measurements from 8-port measurements between victim
pair 602 (port1, port2, port7 and port8) and the first culprit pair
603 (port3, port4, port 9 and port10). These measurements provide
information about the coupling that occurs to the victim pair due
to the first culprit pair (i.e., culprit1). Regarding the matrix
700b of FIG. 7B, shaded portions (e.g., 702b) represent s-parameter
measurements from 8-port measurements between victim pair 602
(port1, port2, port7 and port8) and the second culprit pair 604
(port5, port6, port11 and port12). The matrix 700c of FIG. 7C
results from combining the matrices shown in FIGS. 7A and 7B. Note
that some s-parameters (e.g., 3,5) have not been measured.
S-parameter measurements not representing primary coupling effects
(primary coupling effects corresponding to coupling effects between
the victim pair and a culprit pair) may be ignored in some
embodiments. The s-parameter 3,5, for example, do not represent a
primary coupling effect since this parameter involves coupling
between ports corresponding to culprit pairs (culprit pair1 and
culprit pair2). In the embodiments described herein, primary
coupling effects are of interest, and thus it has been determined
that experimentally, it is of no significance for the purpose of
adequately characterizing the 12-port DUT (606) to make
measurements of these parameters. This determination of
significance can also be performed in the context of a
cost-benefits analysis. For example, although such measurements may
be taken, in some instances, the benefits of taking all s-parameter
measurements may be outweighed by the cost in time and money in
performing the measurements and processing. In some embodiments,
the engineer or designer may determine that he or she cannot ignore
the effect of those un-measured terms.
[0051] FIG. 8 is a schematic diagram that illustrates a 16-port DUT
806 with a victim pair 802 and three culprit pairs (803, 804, and
805). With continued reference to FIG. 8, FIGS. 9A-9D are schematic
diagrams of matrices (900a-900d) that illustrate matrix processing
as implemented by the method 110b (FIG. 5). Similar to the
processing shown in FIGS. 7A-7C, 8-port s-parameter measurements
are taken, with increasing coverage of the s-parameters as shown in
matrices 900a-900c (FIGS. 9A-9C). These s-parameter measurements
are combined in similar manner to that described above, resulting
in the coverage shown in matrix 900d of FIG. 9D. Again, not all
s-parameters are covered, but that is acceptable for this
embodiment as experimentally confirmed.
* * * * *